Ch. 5: Banks in Capital Markets 195
of successful capital market financing from choosing an investment bank. Two key im-
plications of the model are that universal banks will likely be selected as the underwriter
when economies of scope are large, and underwriting allows banks to form and main-
tain strong relationships in multiple product lines, with firms that use universal banks
for underwriting more likely to use the same universal bank for bank lending services.
Puri (1999)models the trade-off between commercial banks’ potential to be better
certifiers of firm value and the conflict of interest that can arise from the bank mis-
representing the value of a firm’s securities in order to use the proceeds to repay bank
loans. The formal model is a repeated game where investors are rational and update
their beliefs about banks given the last period action, which allows reputation concerns
to be captured. Commercial banks know if the firm is good or bad due to previous loan
monitoring activities. Commercial banks can underwrite bad firms in order to pay down
pre-existing claims, but if investors observe this action, then this observation will reduce
the reputation of the bank and its future profits. Investment banks, which do not know
the quality of the firm, can incur an investigative cost to determine the true quality of the
firm. If they choose not to investigate and subsequently underwrite a low quality firm,
the investment bank will suffer through reputation loss. The trade-offs that each under-
writer faces determine equilibrium strategies and the prices that the market assigns to
underwritten securities. A key result is that commercial banks are likely to obtain better
prices for underwritten securities than investment banks when the costs of information
production are high, as might be seen in junior and informational sensitive securities.^6
These results help provide theoretical underpinnings for many of the results in the em-
pirical literature. This paper also examines if it is possible for commercial banks and
investment banks to coexist in equilibrium and derives sufficient conditions for coexis-
tence in which the level of rent extraction and the relative underwriting fees adjust so
that firms are indifferent between going to commercial banks and investment banks. In
Section4, we survey the empirical evidence on competition between commercial banks
and investment banks, including the literature on underwriting fees.
- Empirical evidence on conflicts of interest
Like the theoretical literature, the empirical studies also examine the benefits and costs
of universal banking. Much of the focus of the empirical literature is on the effect of
bank’s lending, and the private information contained therein, on commercial banks’
ability to certify firm value in the presence of potential conflicts of interest. In this sec-
tion, we provide an overview of this literature. The literature examines these issues over
multiple periods of time in the United States. In Section3.1, we provide a summary
(^6) Of course, there are other scenarios where an investment bank can achieve higher prices than a commercial
bank, such as when the costs of investigation for the investment bank are sufficiently small, or when both types
of underwriters are perceived to have low reputations.